a
b Figure 1. Optical image of study area provided by Google
Earth a and the average amplitude image of 25 ENVISAT ASAR images b
3. METHODS
There are two parts contained in the small baseline time series InSAR technique, including linear deformation retrieval and
non-linear deformation retrieval. Let us start our analysis by considering N SAR images acquired at the ordered times. Based
on the principle of small spatial and temporal baselines, we can generate M interferograms. Before linear deformation retrieval,
high coherence point targets are selected according to pixel
’s coherence stability by setting a suitable coherence threshold for
the mean coherence image. Based on these point targets, differential phase are connected with Delaunay triangulation.
Thus the phase slope between two neighboring points
, , ,
m m
n n
x y
x y on an edge can be expressed as
Table 1. List of the perpendicular and temporal baselines of 25 ENVISAT ASAR images
4 ,
, ,
, ,
, 4
, ,
sin ,
, ,
, ,
,
dif m
m n
n i
i m
m n
n i
m m
n n
i i
m m
n n
m m
n n
m m
n n
x y
x y T T
v x y
v x y b T
x y
x y r T
x y
x y x
y x y
n x y
n x y
1
where
and
v
are the height error and linear velocity; ,
m m
x y
and
,
n n
x y are pixel position coordinates;
i
T
is the time baseline of the
i
th interferogram; the nonlinear component of velocity;
the atmospheric phase artefacts; and
n
the decorrelation noise. It is assumed that, within the atmospheric
correlation range 1-3km, the atmospheric phases are equal, thus the atmospheric components can be neglected. For the linear
deformation velocity and height error are constants, thus the above phase slope can be modelled as
Date Perpendicular
baseline m Temporal
Baseline day 1
2006-1-22 2
2006-2-26 696
35 3
2006-5-7 1154
105 4
2006-9-24 252
245 5
2006-10-29 629
280 6
2006-12-3 919
315 7
2007-3-18 1330
420 8
2007-4-22 959
455 9
2007-5-27 946
490 10
2007-7-1 981
525 11
2007-8-5 904
560 12
2007-9-9 1171
595 13
2007-10-14 824
630 14
2007-12-23 606
700 15
2008-5-11 986
840 16
2008-7-20 1157
910 17
2008-12-7 845
1050 18
2009-1-11 1017
1085 19
2009-2-15 1060
1120 20
2009-3-22 1405
1155 21
2009-4-26 990
1190 22
2009-8-9 918
1295 23
2010-6-20 1066
1610 24
2010-7-25 753
1645 25
2010-8-29 915
1680
2015 International Workshop on Image and Data Fusion, 21 – 23 July 2015, Kona, Hawaii, USA
This contribution has been peer-reviewed. doi:10.5194isprsarchives-XL-7-W4-181-2015
182
mod mod
mod
4 ,
, ,
, ,
4 ,
sin
el m
m n
n i
i el
i el
i i
x y
x y T T
v m n
b T m n
r T
2
where
v
are
velocity and height error increments, respectively. They can be retrieved by maximizing the following
Ensemble Phase Coherence EPC Ferretti, 2000
mod mod
, ,
, ,
1 ,
, ,
exp ,
, ,
,
M dif
m m
n n
i el
m m
n n
i el
m m
n n
i
j x
y x y T
x y
x y M
x y
x y T
3 where
j
is the imaginary unit,
M
is the number of interferograms. When the maximum EPC is close to 1, the
velocity and height error increments are close to the real value. Then, the linear velocity and height error on each point target is
obtained by integrating
mod el
v
and
mod el
with EPC over 0.7 from a starting reference point.
To retrieve non-linear deformation, it is necessary to calculate the model phase contributed by linear deformation and height
errors. By subtracting the model phase from differential phase, we get residual phases, which mainly include atmospheric phase,
non-linear deformation component and phase noises. Phase noises can be reduced by spatial low pass filtering. Atmospheric
phase and non-linear deformation can be separated according to their different frequency characteristics in temporal and spatial
domains.
4. RESULTS AND DISCUSSION
Kategori